BENCH: asv csv reading benchmarks no longer read StringIO objects off the end

* benchmarks for read_csv() now properly rewind StringIO objects prior to
reading them in; previously, all iterations of an asv repeat timing run
would read in no data because the StringIO object was pointing to its end after
the first iteration--setup() only runs between repeats, not iterations within
repeats of timeit

Author: tylerjereddy

asv_bench/benchmarks/io/csv.py

@@ -69,6 +69,7 @@ def setup(self, infer_datetime_format, format):
         self.data = StringIO('\n'.join(rng.strftime(dt_format).tolist()))
 
     def time_read_csv(self, infer_datetime_format, format):
+        self.data.seek(0)
         read_csv(self.data, header=None, names=['foo'], parse_dates=['foo'],
                  infer_datetime_format=infer_datetime_format)
 
@@ -108,12 +109,15 @@ def setup(self):
         self.data2 = StringIO('\n'.join(arr.astype(str).tolist()))
 
     def time_read_uint64(self):
+        self.data1.seek(0)
         read_csv(self.data1, header=None, names=['foo'])
 
     def time_read_uint64_neg_values(self):
+        self.data2.seek(0)
         read_csv(self.data2, header=None, names=['foo'])
 
     def time_read_uint64_na_values(self):
+        self.data1.seek(0)
         read_csv(self.data1, header=None, names=['foo'],
                  na_values=self.na_values)
 
@@ -149,6 +153,7 @@ def setup(self):
         self.s_data = StringIO('\n'.join(data))
 
     def time_comment(self):
+        self.s_data.seek(0)
         read_csv(self.s_data, comment='#', header=None, names=list('abc'))
 
 
@@ -167,10 +172,12 @@ def setup(self, sep, decimal, float_precision):
         self.s_data = StringIO(data)
 
     def time_read_csv(self, sep, decimal, float_precision):
+        self.s_data.seek(0)
         read_csv(self.s_data, sep=sep, header=None, names=list('abc'),
                  float_precision=float_precision)
 
     def time_read_csv_python_engine(self, sep, decimal, float_precision):
+        self.s_data.seek(0)
         read_csv(self.s_data, sep=sep, header=None, engine='python',
                  float_precision=None, names=list('abc'))
 
@@ -209,9 +216,11 @@ def setup(self):
         self.s_data = StringIO(data)
 
     def time_multiple_date(self):
+        self.s_data.seek(0)
         read_csv(self.s_data, sep=',', header=None,
                  names=list(string.digits[:9]), parse_dates=[[1, 2], [1, 3]])
 
     def time_baseline(self):
+        self.s_data.seek(0)
         read_csv(self.s_data, sep=',', header=None, parse_dates=[1],
                  names=list(string.digits[:9]))